Re: [PATCHv2 0/7] direct-io: even more flexible io vectors

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Aug 07, 2025 at 05:20:17PM -0600, Keith Busch wrote:
> Sure. I wrote up some for blktest, and the same test works as-is for
> filesystems too. Potential question: where do such programs go
> (xfstests, blktests, both, or some common place)?

We currently have no good way to share tests between xfstest and
blktests, so I think it would require duplicating the helper programs.

> I tested on loop, nvme, and virtio-blk, both raw block (blktests) and
> xfs (fstests). Seems fine.

Cool.  I'd like the hear from the other XFS folks if the possibility
of easily introducing preallocated space (that's what it looks like on
disk) for writes failed because of wrong alignment is fine.  Given that
the same user could introduce them using fallocate it's definitively not
a security issue, but also rather unusual and unexpected.  And from the
other file system maintainers if they have similar issues.





[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux